The present invention relates to a method for re-tracing the end point of an image and, more particularly, to a method for re-tracing the end point, or a terminal point, of the image, which can be utilized for image processing devices, such as a CAD (computer-aided design) machine and so on, preferably in semi-automatically tracing line images of image data and which is so adapted as to continuously re-trace the end point of the image which follows the end point of the previously traced image.
Heretofore, the method for the input of a drawing drawn on paper into an image processing system generally comprises inputting of the drawing as image data by an image scanner or the like, displaying the image data on a display screen by an image processing system, and tracing the image by allowing the operator to manually specify the tracing points on the image with a pointing device or the like while viewing the images on the display screen.
For example, the method for the input of a drawing in a conversational mode as disclosed in Japanese Patent Laid-open Publication (kokai) No. 269,276/1987 involves entering the drawing into a computer system as image data through a scanner, superimposing the resulting image data on center-line image data on the display screen, and entering the drawing as vector data representing the drawing with coordinate points by specifying necessary positions of the drawing on the display screen with a coordinate pointer.
For example, when the image data is converted into graphic data such as vector data, etc. as described hereinabove, central data on the center or the center line of the image is given, data on necessary characteristic points of the continuous image, such as end points, branch points, corner points, etc. is selected and specified from the center line of the continuous central data, and such data is entered. In this case, in order to enter the characteristic points, the central data is superimposed on the image data on the display screen to allow the operator to judge the characteristic points with ease, and the characteristic points are entered manually by the operator. Hence, this input method suffers from the disadvantage that the display screen becomes so complex that the operator cannot select the necessary data with high efficiency.
On the other hand, there has been developed an image processing device with a semi-automatically image-trace processing function that traces image points in a semi-automatic way, determines the characteristic points, and enters graphic data of the characteristic points. For such a semi-automatically image-trace processing function of the image processing device, only one line image as an object for conversion into vector data can be traced by tracing the line images upon one instruction to start tracing.
Therefore, when the line images, such as contour lines of a map graphic, etc., are traced in a semi-automatic way by taking advantage of the semi-automatically image-trace processing function of the image processing device, the continuity of the line images (a continuous pixel sequence) of the contour lines is disconnected, when there is an image, such as numeral font data which represents height between the line images and which does not comprise any line image, so that the tracing cannot be performed in a continuous fashion and the line images cannot be converted into one vector data. In other words, even if the line image composed originally of one line would be divided into plural line images, it cannot be converted into one vector data.